Image processing method and device, electronic equipment, storage medium and program product

Abstract

The invention provides an image processing method and device, electronic equipment, a storage medium and a program product, and relates to the technical field of image processing. The method comprises the following steps: performing visual feature extraction on a to-be-processed image to obtain a fusion feature vector; the fusion feature vector is input into a pre-constructed decision tree model to obtain an image enhancement parameter set, the decision tree model comprises a node splitting threshold, and the node splitting threshold is dynamically adjusted according to historical operation data of a user; and processing the to-be-processed image according to the image enhancement parameter set. According to the method, the problems that a traditional image needs manual parameter adjustment processing and cannot be self-adapted according to user habits are solved, the enhanced parameters automatically fit user preferences by analyzing historical operation data of the user and adjusting the node splitting threshold value of the decision tree in real time, and the image processing efficiency is remarkably improved.

Description

Technical Field

[0001] This disclosure relates to the field of image processing technology, and in particular to an image processing method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] Image enhancement technology is widely used in medical image analysis, mobile photography, and other fields to improve image quality and visual effects. Traditional image enhancement methods typically rely on manual adjustment of parameters such as brightness and contrast, or use fixed algorithms with preset parameters, lacking the ability to learn from user habits. When users need to make personalized adjustments for different scenarios, they must repeatedly modify parameter settings manually, which is cumbersome and inefficient. Especially in professional scenarios such as medical imaging diagnosis, different physicians have significantly different preferences for image enhancement parameters, and traditional methods cannot adapt to user habits, leading to reduced work efficiency and limited user experience. Therefore, how to make image enhancement parameters automatically adapt to user habits and reduce manual intervention has become a key technical issue in improving the intelligence level of image processing.

[0003] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] This disclosure provides an image processing method, apparatus, electronic device, storage medium, and program product, which at least to some extent overcomes the problem that traditional image enhancement methods require manual parameter adjustment and cannot adapt to user habits.

[0005] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0006] According to one aspect of this disclosure, an image processing method is provided, comprising: extracting visual features from an image to be processed to obtain a fused feature vector; inputting the fused feature vector into a pre-constructed decision tree model to obtain an image enhancement parameter set, the decision tree model including a node splitting threshold, the node splitting threshold being dynamically adjusted according to user historical operation data; and completing the processing of the image to be processed according to the image enhancement parameter set.

[0007] In some embodiments, visual feature extraction of the image to be processed to obtain a fused feature vector includes: extracting shallow texture features of the image to be processed through shallow convolutional layers of a convolutional neural network; extracting deep semantic features of the image to be processed through deep convolutional layers of the convolutional neural network; and generating the fused feature vector based on the shallow texture features and the deep semantic features.

[0008] In some embodiments, the method further includes: performing online pruning on the decision tree model, wherein the online pruning operation includes removing decision paths whose call frequency is lower than a preset threshold; storing the decision tree node data after the online pruning operation is performed, and generating a lightweight decision tree model.

[0009] In some embodiments, inputting the fused feature vector into a pre-built decision tree model to obtain an image enhancement parameter set includes: collecting user historical operation data, wherein the user historical operation data is user adjustment behavior data of image enhancement parameters; calculating dynamic adjustment coefficients based on the user historical operation data based on a time decay factor; updating the decision tree node splitting thresholds according to the dynamic adjustment coefficients to generate an image enhancement parameter set matching user preferences.

[0010] In some embodiments, the dynamic adjustment coefficient is determined by the following formula:

[0011] θt=αθt-1+(1-α)Δθt

[0012] Where θt represents the adjustment coefficient at the current moment, α represents the time decay factor, θt-1 represents the adjustment coefficient at the previous moment, and Δθt represents the parameter offset generated by the current user operation.

[0013] In some embodiments, storing decision tree node data after online pruning operations to generate a lightweight decision tree model includes: performing differential encoding compression on the decision tree node data after online pruning operations; and storing the topology of the decision tree model separately from the decision tree node data to achieve updating of the decision tree model.

[0014] According to another aspect of this disclosure, an image processing apparatus is also provided, comprising: a feature extraction module for extracting visual features from an image to be processed to obtain a fused feature vector; a parameter generation module for inputting the fused feature vector into a pre-constructed decision tree model to obtain an image enhancement parameter set, the decision tree model including a node splitting threshold, the node splitting threshold being dynamically adjusted according to user historical operation data; and an image processing module for processing the image to be processed according to the image enhancement parameter set.

[0015] According to another aspect of this disclosure, an electronic device is also provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the image processing method described in any of the preceding claims by executing the executable instructions.

[0016] According to another aspect of this disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the image processing method described in any of the preceding claims.

[0017] According to another aspect of this disclosure, a computer program product is also provided, including a computer program that, when executed by a processor, implements the image processing method of any one of the above.

[0018] The image processing methods, apparatus, electronic devices, storage media, and program products provided in the embodiments of this disclosure extract visual features from the image to be processed to obtain a fused feature vector; input the fused feature vector into a pre-constructed decision tree model to obtain a set of image enhancement parameters, the decision tree model including node splitting thresholds, the node splitting thresholds being dynamically adjusted according to the user's historical operation data; and complete the processing of the image to be processed based on the set of image enhancement parameters. The embodiments of this disclosure overcome the problem that traditional image processing requires manual parameter tuning and cannot adapt to user habits. By analyzing the user's historical operation data, the decision tree node splitting thresholds are adjusted in real time, allowing the enhancement parameters to automatically match user preferences, significantly improving image processing efficiency and personalization.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0021] Figure 1 This diagram illustrates an exemplary application system architecture in an embodiment of the present disclosure.

[0022] Figure 2 This diagram illustrates a flowchart of an image processing method according to an embodiment of the present disclosure;

[0023] Figure 3 This diagram illustrates a flowchart of yet another image processing method according to an embodiment of the present disclosure;

[0024] Figure 4 This diagram illustrates a flowchart of yet another image processing method according to an embodiment of the present disclosure;

[0025] Figure 5 This diagram illustrates a flowchart of yet another image processing method according to an embodiment of the present disclosure;

[0026] Figure 6 This diagram illustrates an image processing apparatus according to an embodiment of the present disclosure.

[0027] Figure 7 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0028] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0029] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0030] The specific implementation methods of the embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0031] Figure 1 A schematic diagram of an exemplary application system architecture to which the image processing methods of the embodiments of this disclosure can be applied is shown. For example... Figure 1 As shown, the system architecture may include terminal device 101, network 102 and server 103.

[0032] Network 102 is a medium used to provide a communication link between terminal device 101 and server 103, and can be a wired network or a wireless network.

[0033] Optionally, the aforementioned wireless or wired networks use standard communication technologies and / or protocols. The network is typically the Internet, but can also be any network, including but not limited to Local Area Networks (LANs), Metropolitan Area Networks (MANs), Wide Area Networks (WANs), mobile, wired or wireless networks, private networks, or any combination of virtual private networks. In some embodiments, technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc., are used to represent data exchanged over the network. Furthermore, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Networks (VPNs), and Internet Protocol Security (IPSec) can be used to encrypt all or some links. In other embodiments, custom and / or dedicated data communication technologies can be used to replace or supplement the aforementioned data communication technologies.

[0034] Terminal device 101 can be various electronic devices, including but not limited to smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, wearable devices, augmented reality devices, virtual reality devices, etc.

[0035] Optionally, the client of the application installed on different terminal devices 101 may be the same, or the client of the same type of application based on different operating systems. Depending on the terminal platform, the specific form of the application client may also be different; for example, the application client may be a mobile client, a PC client, etc.

[0036] Server 103 can be a server that provides various services, such as a backend management server that supports the device operated by the user using terminal device 101. The backend management server can analyze and process received requests and other data, and feed the processing results back to the terminal device.

[0037] Optionally, the server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0038] Those skilled in the art will know that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative; any number of terminal devices, networks, and servers can be included depending on actual needs. This disclosure does not limit the scope of the embodiments.

[0039] Under the above system architecture, this disclosure provides an image processing method that can be executed by any electronic device with computing capabilities.

[0040] In some embodiments, the image processing method provided in this disclosure can be executed by a terminal device of the system architecture described above; in other embodiments, the image processing method provided in this disclosure can be executed by a server in the system architecture described above; in still other embodiments, the image processing method provided in this disclosure can be implemented by the terminal device and the server in the system architecture described above through interaction.

[0041] Figure 2 A flowchart of an image processing method according to an embodiment of this disclosure is shown, such as Figure 2 As shown, the image processing method provided in this embodiment includes the following steps:

[0042] S202, Visual feature extraction is performed on the image to be processed to obtain a fused feature vector.

[0043] It should be noted that the image to be processed can be any type of image requiring parameter adjustment, such as medical images, photos taken by mobile devices, or security surveillance footage. Visual feature extraction refers to extracting multi-level visual information from the image to be processed in order to capture the key details and overall structure of the image. Specifically, feature extraction networks can be used for visual feature extraction. The fused feature vector is a comprehensive representation vector generated by combining visual features at different levels and scales.

[0044] S204. The fused feature vector is input into a pre-built decision tree model to obtain a set of image enhancement parameters. The decision tree model includes a node splitting threshold, which is dynamically adjusted based on the user's historical operation data.

[0045] It should be noted that the pre-built decision tree model refers to a decision tree structure pre-trained based on historical data or general rules before the image processing method is deployed. The decision tree model defines the mapping rules from input features (fused feature vectors) to output parameters (a set of image enhancement parameters). Specifically, the decision tree model uses the CART (Classification and Regression Trees) algorithm, optimizing the feature-parameter mapping relationship through a hybrid branching strategy. It uses a large amount of labeled image data as the training set to learn the mapping relationship between feature vectors and optimal enhancement parameters; each decision node has an initial splitting threshold set according to a preset parameter range.

[0046] The image enhancement parameter set is the set of enhancement operation instructions output by the decision tree model, used to adjust the visual attributes of the image, specifically one or more of the following: brightness, contrast, saturation, etc. The node splitting threshold is a critical value used in the decision tree model to determine the branching direction of feature vectors. When feature vectors are input into the decision tree model, the data is assigned to the left or right subtree based on the node splitting threshold.

[0047] It is important to clarify that the node splitting threshold in this embodiment is dynamically adjusted in real time based on the user's historical operation data. This real-time updating of the decision tree node splitting threshold makes the model output more closely match user preferences. By dynamically adjusting the threshold, this embodiment allows the decision tree model to continuously learn user operating habits. For example, if a photographer prefers a high-contrast style, the decision tree model will automatically increase the splitting threshold of contrast-related nodes. This avoids the tedious manual adjustment of parameters in traditional methods, improving processing efficiency; dynamic adjustment only requires updating the threshold parameters, eliminating the need to retrain the entire model, thus reducing model training overhead.

[0048] Furthermore, this embodiment employs a two-stage decoupled architecture, separating feature extraction from parameter decision-making. The feature extraction stage uses a deep convolutional network to extract multi-scale image features (shallow texture + deep semantics), ensuring comprehensive feature representation. A feature-parameter mapping model is established based on an improved hybrid decision tree, enabling interpretable enhancement logic. Traditional end-to-end black-box models (such as CNNs that directly output parameters) lack interpretability, while this embodiment visualizes user preference logic through decision tree paths. The modular design supports independent upgrades of the feature extractor (e.g., replacement with ResNet, ViT, etc.), adapting to different scenario requirements.

[0049] S206, complete the processing of the image to be processed based on the set of image enhancement parameters.

[0050] It's important to note that the image enhancement parameter set output by the decision tree model is transformed into specific image processing operations, ultimately generating an enhanced image that matches user preferences. The core objective is to eliminate the efficiency bottleneck of traditional methods that rely on manual adjustments through automated parameter execution, while ensuring that the processing results accurately meet user needs. By adjusting the image to be processed through the enhancement parameter set, subjective biases from manual operation are avoided. From feature extraction to parameter execution, no manual intervention is required throughout, resulting in a significant improvement in model output efficiency.

[0051] This disclosure provides an image processing method that adaptively generates image enhancement parameters using a dynamic decision tree model. The core objective is to address the shortcomings of traditional methods that rely on manual parameter tuning and cannot adapt to user operating habits. This disclosure analyzes historical user operation data and adjusts the splitting threshold of decision tree nodes in real time, automatically aligning enhancement parameters with user preferences and significantly improving image processing efficiency and personalization.

[0052] In some embodiments of this disclosure, such as Figure 3 As shown, visual feature extraction of the image to be processed to obtain the fused feature vector also includes the following steps:

[0053] S302 extracts shallow texture features of the image to be processed through shallow convolutional layers of a convolutional neural network.

[0054] It should be noted that shallow convolutional layers refer to the low-level convolutional layers (such as layers 1-3) near the input in a convolutional neural network (CNN). Their core function is to capture local details and basic texture information of the image. In some embodiments of this disclosure, the first three layers of a classic CNN architecture such as ResNet can be used as shallow feature extractors; the kernel size is typically 3×3 or 5×5, and the stride is set to 1 to preserve high-resolution details. The shallow texture features of the image to be processed include image edges, texture details, noise distribution, etc. For example, in medical images, shallow texture features can clearly present tissue boundaries and small lesions. In some embodiments, shallow convolutional layers use 3×3 small kernel convolutions to avoid loss of detail information, which is particularly suitable for the extraction of small nodules in medical images.

[0055] S304 extracts deep semantic features of the image to be processed through deep convolutional layers of a convolutional neural network.

[0056] It's important to note that deep convolutional layers refer to higher-level convolutional layers (such as layers 4-6) near the output in a CNN. These layers capture global semantic information of the image by expanding the receptive field. Deep semantic features include region contrast, object contours, and brightness distribution. For example, in photographic images, deep features can identify the lighting characteristics of semantic objects such as the sky and faces. Deep semantic features capture global context through dilated convolutions, addressing the problem of traditional methods' poor adaptability to complex lighting scenes.

[0057] S306 generates a fused feature vector based on shallow texture features and deep semantic features.

[0058] It should be noted that the fused feature vector is a comprehensive representation generated by combining shallow texture features and deep semantic features, which simultaneously preserves local details and global structural information. Feature fusion employs channel concatenation and 1×1 convolution compression, which reduces the number of feature channels and lowers computational complexity, enabling real-time processing on GPU devices at the 15ms level.

[0059] This embodiment constructs an image representation system that takes into account both details and semantics by collaboratively extracting shallow and deep features and fusing them across scales. This provides a high-information-density input for subsequent dynamic decision trees, breaking through the limitations of traditional single-layer features and achieving a dual enhancement effect of "local accuracy + global intelligence".

[0060] In some embodiments of this disclosure, such as Figure 4 As shown, the image processing method also includes the following steps:

[0061] S402, Perform online pruning on the decision tree model. The online pruning operation includes removing decision paths whose call frequency is lower than a preset threshold.

[0062] It should be noted that the online pruning operation described in this embodiment refers to dynamically removing inefficient decision paths during the operation of the decision tree model, in order to optimize the model structure and reduce computational resource consumption. Its core purpose is to ensure the real-time processing performance of edge devices by reducing the size of the decision tree in real time. Specific online pruning operations may include removing decision paths whose call frequency is below a preset threshold.

[0063] S404 stores the decision tree node data after the online pruning operation is completed, generating a lightweight decision tree model.

[0064] It should be noted that the decision tree model obtained by storing the decision tree node data after performing online pruning is significantly lighter than the decision tree model without pruning. This lightweight design is reflected in...

[0065] By reducing the number of nodes, specifically by removing inefficient decision paths with a call frequency below a preset threshold, the number of model nodes is significantly reduced. This reduction in the number of nodes compresses storage space and lowers memory usage, thus achieving a lightweight decision tree model.

[0066] It should be added that the lightweight deployment solution achieves the following resource optimizations:

[0067] Memory usage: The model size is reduced by 83% compared to traditional DNNs, and memory usage is reduced by 58%;

[0068] Computational efficiency: A single GPU (Graphics Processing Unit) device supports real-time processing at the 15ms level, reducing CPU utilization by 31%;

[0069] Compatibility: Adaptable to various scenarios including mobile phones (ARM (Advanced RISC Machines) architecture), medical devices (embedded GPUs), and security cameras (low-power chips).

[0070] Specifically, storing the decision tree node data after performing online pruning operations and generating a lightweight decision tree model can include the following steps:

[0071] Differential encoding compression is performed on the decision tree node data after online pruning;

[0072] The topology of the decision tree model is stored separately from the decision tree node data to enable updates to the decision tree model.

[0073] It should be noted that differential coding compression refers to significantly reducing the storage space occupied by decision tree models by calculating the parameter differences between adjacent nodes and performing entropy coding, thus solving the storage bottleneck problem of traditional decision tree models when deployed on edge devices. Specifically, differential coding compression can be achieved through difference matrix generation and entropy coding compression, which can be implemented using Huffman coding and run-length encoding.

[0074] Topology refers to the hierarchical relationship between nodes in a decision tree, while node data refers to the parameter values ​​of each node (such as brightness threshold and contrast gain). The separate storage strategy decouples the structure and parameters, enabling efficient updates and version management of the decision tree model.

[0075] Some embodiments of this disclosure achieve compressed model size, overcoming storage limitations of edge devices; optimize data decoding and loading processes to ensure real-time performance; and support dynamic incremental updates to adapt to various scenario requirements.

[0076] In some embodiments of this disclosure, such as Figure 5As shown, inputting the fused feature vector into a pre-built decision tree model to obtain the image enhancement parameter set also includes the following steps:

[0077] S502 collects user historical operation data, which is the user's behavior data on adjusting image enhancement parameters.

[0078] It should be noted that the user's historical operation data is the user's behavior data on adjusting image enhancement parameters, that is, the user's historical parameter tuning data for images. The purpose of collecting the user's historical operation data is to build a user preference profile to drive the dynamic optimization of the decision tree model.

[0079] Specific historical operation data may include one or more of the following:

[0080] Adjustment type: The category of parameters modified by the user (such as brightness, contrast, saturation);

[0081] Adjustment range: parameter correction amount (e.g., if the brightness is adjusted from 80 to 100, the parameter correction amount is 20);

[0082] Timestamp: The point in time when the operation occurred, used to calculate the time decay weight.

[0083] In addition, after collecting historical user operation data, a data storage step can be included. A sliding time window is used to store user operation data, with data exceeding the window period automatically weighted and decayed.

[0084] S504 calculates and generates dynamic adjustment coefficients based on the user's historical operation data using a time decay factor.

[0085] It should be noted that the dynamic adjustment coefficient is a quantitative indicator obtained by weighting historical operation data using a time decay factor. It reflects the dynamic trend of user preferences and can be determined using the following formula:

[0086] θt=αθt-1+(1-α)Δθt (1)

[0087] Where θt represents the adjustment coefficient at the current moment, α represents the time decay factor, θt-1 represents the adjustment coefficient at the previous moment, and Δθt represents the parameter offset generated by the current user operation.

[0088] In other embodiments of this disclosure, multi-dimensional feature calculations can be added to the above formula (1), such as adding a visual sensitivity matrix. The visual sensitivity matrix is ​​a weight map generated by analyzing the frequency of parameter adjustment of different regions of the image by the user, which quantifies the user's visual attention to a specific region. The sensitivity matrix can make local image enhancement more accurate.

[0089] S506, Update the decision tree node splitting threshold according to the dynamic adjustment coefficient, and generate a set of image enhancement parameters that match the user's preferences.

[0090] It should be noted that the dynamic adjustment coefficients are mapped to the decision tree node splitting threshold to achieve real-time alignment between model parameters and user preferences.

[0091] The embodiments disclosed herein generate dynamically adjustable coefficients that quantify changes in user preferences, making the final set of image enhancement parameters output by the decision tree model more closely match the user's recent operating habits, balancing the user's long-term habits and short-term needs. Compared to traditional static decision tree schemes, it can update user profiles more dynamically, shorten the adaptation period for new users to use the decision tree model, and its storage cost is more lightweight than traditional models.

[0092] Based on the same inventive concept, this disclosure also provides an image processing apparatus, as described in the following embodiments. Since the principle by which this apparatus solves the problem is similar to that of the method embodiments described above, the implementation of this apparatus embodiment can refer to the implementation of the method embodiments described above, and repeated details will not be repeated.

[0093] Figure 6 This diagram illustrates an image processing apparatus according to an embodiment of the present disclosure, such as... Figure 6 As shown, the device includes:

[0094] Feature extraction module 601 is used to extract visual features from the image to be processed and obtain a fused feature vector;

[0095] The parameter generation module 602 is used to input the fused feature vector into a pre-built decision tree model to obtain a set of image enhancement parameters. The decision tree model includes a node splitting threshold, which is dynamically adjusted according to the user's historical operation data.

[0096] The image processing module 603 is used to process the image to be processed according to the set of image enhancement parameters.

[0097] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above method embodiments. It should also be noted that the above modules, as part of an apparatus, can be executed in a computer system such as a set of computer-executable instructions.

[0098] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0099] The following reference Figure 7 To describe an electronic device 700 according to such an embodiment of the present disclosure. Figure 7 The electronic device 700 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0100] like Figure 7 As shown, the electronic device 700 is manifested in the form of a general-purpose computing device. The components of the electronic device 700 may include, but are not limited to: at least one processing unit 710, at least one storage unit 720, and a bus 730 connecting different system components (including storage unit 720 and processing unit 710).

[0101] The storage unit stores program code that can be executed by the processing unit 710, causing the processing unit 710 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit 710 can perform the following steps of the above method embodiments: extracting visual features from the image to be processed to obtain a fused feature vector; inputting the fused feature vector into a pre-constructed decision tree model to obtain an image enhancement parameter set, the decision tree model including a node splitting threshold, the node splitting threshold being dynamically adjusted according to the user's historical operation data; and completing the processing of the image to be processed according to the image enhancement parameter set.

[0102] Storage unit 720 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 7201 and / or a cache memory unit 7202, and may further include a read-only memory unit (ROM) 7203.

[0103] The storage unit 720 may also include a program / utility 7204 having a set (at least one) program module 7205, such program module 7205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0104] Bus 730 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0105] Electronic device 700 can also communicate with one or more external devices 740 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 700, and / or with any device that enables electronic device 700 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 750. Furthermore, electronic device 700 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 760. As shown, network adapter 760 communicates with other modules of electronic device 700 via bus 730. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0106] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0107] In particular, according to embodiments of this disclosure, the process described above with reference to the flowchart can be implemented as a computer program product, which includes a computer program that, when executed by a processor, implements the methods described in the above embodiments.

[0108] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. A program product capable of implementing the methods described above is stored thereon. In some possible implementations, various aspects of this disclosure may also be implemented as a program product including program code, which, when run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0109] More specific examples of computer-readable storage media in this disclosure may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0110] In this disclosure, a computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting a program for use by or in connection with an instruction execution system, apparatus, or device.

[0111] Optionally, the program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0112] In practical implementation, program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0113] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0114] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0115] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0116] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the appended claims.

Claims

1. An image processing method, characterized in that, include: Visual features are extracted from the image to be processed to obtain a fused feature vector; The fused feature vector is input into a pre-constructed decision tree model to obtain a set of image enhancement parameters. The decision tree model includes a node splitting threshold, which is dynamically adjusted based on the user's historical operation data. The image to be processed is processed according to the set of image enhancement parameters.

2. The image processing method according to claim 1, characterized in that, Visual features are extracted from the image to be processed to obtain a fused feature vector, including: The shallow texture features of the image to be processed are extracted through the shallow convolutional layers of the convolutional neural network. The deep semantic features of the image to be processed are extracted through the deep convolutional layers of the convolutional neural network; The fused feature vector is generated based on the shallow texture features and the deep semantic features.

3. The image processing method according to claim 1, characterized in that, The method further includes: Perform online pruning on the decision tree model. The online pruning operation includes removing decision paths whose call frequency is lower than a preset threshold. Store the decision tree node data after the online pruning operation is performed to generate a lightweight decision tree model.

4. The image processing method according to claim 1 or 3, characterized in that, The fused feature vector is input into a pre-built decision tree model to obtain a set of image enhancement parameters, including: Collect user historical operation data, which is user behavior data on adjusting image enhancement parameters; Dynamic adjustment coefficients are generated by calculating the user's historical operation data based on the time decay factor. The decision tree node splitting threshold is updated based on the dynamically adjusted coefficients to generate a set of image enhancement parameters that match user preferences.

5. The image processing method according to claim 4, characterized in that, The dynamic adjustment coefficient is determined using the following formula: θt=αθt-1+(1-α)Δθt Where θt represents the adjustment coefficient at the current moment, α represents the time decay factor, θt-1 represents the adjustment coefficient at the previous moment, and Δθt represents the parameter offset generated by the current user operation.

6. The image processing method according to claim 3, characterized in that, Store the decision tree node data after online pruning, and generate a lightweight decision tree model, including: Differential encoding compression is performed on the decision tree node data after online pruning; The topology of the decision tree model is stored separately from the decision tree node data to enable the updating of the decision tree model.

7. An image processing apparatus, characterized in that, include: The feature extraction module is used to extract visual features from the image to be processed and obtain a fused feature vector. The parameter generation module is used to input the fused feature vector into a pre-constructed decision tree model to obtain an image enhancement parameter set. The decision tree model includes a node splitting threshold, which is dynamically adjusted according to the user's historical operation data. The image processing module is used to process the image to be processed according to the set of image enhancement parameters.

8. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the image processing method according to any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the image processing method according to any one of claims 1 to 6.

10. A computer program product, comprising: A computer program or instruction, characterized in that, when executed by a processor, the computer program or instruction implements the image processing method according to any one of claims 1 to 6.

Images